[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240327171807.27741-1-xry111@xry111.site>
Date: Thu, 28 Mar 2024 01:18:07 +0800
From: Xi Ruoyao <xry111@...111.site>
To: Huacai Chen <chenhuacai@...nel.org>,
WANG Xuerui <kernel@...0n.name>
Cc: loongarch@...ts.linux.dev,
linux-kernel@...r.kernel.org,
Tiezhu Yang <yangtiezhu@...ngson.cn>,
Xi Ruoyao <xry111@...111.site>
Subject: [PATCH] LoongArch: Select ARCH_HAS_FAST_MULTIPLIER
LA464 and LA664 can do 32-bit/64-bit integer multiplication with a
latency of 4 cycles and a throughput of 2 ops per cycle. It's
comparable to mainstream x86 and arm64 cores, so select
ARCH_HAS_FAST_MULTIPLIER like them.
It speeds up __sw_hweight32 in lib/hweight.c for about 14% on LA464 and
11% on LA664, and __sw_hweight64 for about 30% on LA464 and 33% on
LA664.
Signed-off-by: Xi Ruoyao <xry111@...111.site>
---
arch/loongarch/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig
index 5a769bb92d7c..d52a95195e7f 100644
--- a/arch/loongarch/Kconfig
+++ b/arch/loongarch/Kconfig
@@ -16,6 +16,7 @@ config LOONGARCH
select ARCH_HAS_ACPI_TABLE_UPGRADE if ACPI
select ARCH_HAS_CPU_FINALIZE_INIT
select ARCH_HAS_CURRENT_STACK_POINTER
+ select ARCH_HAS_FAST_MULTIPLIER
select ARCH_HAS_FORTIFY_SOURCE
select ARCH_HAS_KCOV
select ARCH_HAS_NMI_SAFE_THIS_CPU_OPS
--
2.44.0
Powered by blists - more mailing lists